Molten salts promoting the “controlled carbonization” of waste polyesters into hierarchically porous carbon for high-performance solar steam evaporation

Boyi Zhang a, Changyuan Song a, Chang Liu a, Jiakang Min b, Jalal Azadmanjiri c, Yunxia Ni a, Ran Niu d, Jiang Gong *a, Qiang Zhao *a and Tao Tang *e
aKey Laboratory of Material Chemistry for Energy Conversion and Storage, Ministry of Education, Hubei Key Laboratory of Material Chemistry and Service Failure, School of Chemistry and Chemical Engineering, Huazhong University of Science and Technology, Wuhan 430074, China. E-mail: gongjiang@hust.edu.cn; zhaoq@hust.edu.cn
bDepartment of Materials Science & Engineering, National University of Singapore, 9 Engineering Drive 1, 117576 Singapore, Singapore
cFaculty of Science, Engineering and Technology, Swinburne University of Technology, Hawthorn, Victoria, 3122 Australia
dDepartment of Physics, Cornell University, Ithaca, New York 14853, USA
eState Key Laboratory of Polymer Physics and Chemistry, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun 130022, China. E-mail: ttang@ciac.ac.cn

Received 16th July 2019 , Accepted 21st August 2019

First published on 23rd August 2019


Solar steam generation is emerging as a promising technology to harvest solar energy for diverse applications such as distillation, desalination, and production of freshwater. However, the synthesis of low-cost and high-efficiency photothermal materials remains a challenge for both industrial and academic research. Here, hierarchically porous carbon having an irregular, inter-connected nanoparticle morphology is facilely synthesized through “controlled carbonization” of low-cost waste poly(ethylene terephthalate) (PET) using ZnCl2/NaCl eutectic salts at 550 °C. We prove that ZnCl2 catalyzes the dehydration and decarboxylation of PET to form vinyl-terminated chain fragments and aromatic rings, which subsequently construct a carbon material frame through further cyclization and crosslinking. Meanwhile ZnCl2/NaCl eutectic salts act as porogens to produce mesopores and macropores. The resultant hierarchically porous carbon shows high specific surface area, fast water transportation, high solar absorption efficiency, and low thermal conductivity. Such combined features endow it with a high evaporation rate of 1.68 kg m−2 h−1 under 1 sun irradiation, an energy conversion efficiency of 97%, and a metallic ion removal efficiency of ca. 99.9% for seawater. More importantly, it outperforms the state-of-the-art carbon-based photothermal materials. This work not only reveals the potential of hierarchically porous carbon for application in solar steam generation, but also gives impetus to the sustainable conversion of low-cost waste polyesters into valuable carbon materials by a “controlled carbonization” strategy for energy storage and conversion, environmental remediation, etc.


1. Introduction

With the rapid consumption of fossil fuels, solar energy as a sustainable energy resource has become a promising form of energy to power the development of human society. Converting solar energy to a form that can be directly used or stored is a general approach to utilize solar energy.1–5 Traditionally, evaporation is achieved by heating bulk water. Such a process normally is prone to heat loss and leads to a low solar-to-vapor conversion efficiency. In contrast, solar-driven interfacial evaporation with confined heating at the water–air interface is an emerging and efficient approach in solar energy conversion, due to low heat loss and high photothermal conversion efficiency. It can be used in plenty of applications such as water purification, desalination, and vapor generation. In solar steam generation systems, photothermal materials are some of the most essential elements and should meet several criteria, i.e., broadband sunlight absorbability, open porosity for rapid water molecule transportation, and low thermal conductivity. Until now, a lot of photothermal materials have been developed, e.g., metallic nanoparticles,6–9 biomass-based materials,10,11 and porous polymers.12,13 Porous carbon materials, owing to high porosities, low thermal conductivities, and broad light absorption, show great potential as photothermal materials.14–17 Particularly, a hierarchically porous architecture can form a 3D porous network for efficient water supply and vapor channels, which is crucial for solar steam generation. Unfortunately, to the best of our knowledge, there have been no reports on the application of hierarchically porous carbon in solar steam generation. Besides, chemical and physical activations using acids (e.g., H3PO4) and alkalis (e.g., KOH) are usually employed to introduce nanopores in carbons by etching graphitic layers. However, these methods suffer from the use of harsh acids or alkalis which raises concerns over equipment corrosion, low yields, and energy-intensive procedures. Consequently, the development of easy and cost-effective sustainable synthetic approaches towards hierarchically porous carbon is urgently needed for high-performance solar steam generation.

During the last two decades, millions of tons of waste polymers including plastics and tires were discarded annually into the environment. Since they take hundreds of years to degrade naturally, waste polymers have caused terrible environmental issues (e.g., “White Pollution”) and threatened the ecological environment and human health.18–20 Converting low-cost waste polymers into high value-added carbon nanomaterials not only contributes to sustainable development, but also provides a green approach to prepare porous carbons for diverse applications.21–26 Recently, we proposed the concept of “controlled carbonization” of polymers to synthesize carbons with well-defined structures.27 For noncharring polymers, the backbone chains decompose into low-molecular weight compounds during heating. The key to the controlled carbonization of noncharring polymers is to promote the degradation of polymers into suitable low-molecular weight compounds for the subsequent growth of carbon materials. Williams' group reported catalyzed gasification of waste plastics into carbon nanotubes (CNTs), in which water accelerated the degradation of plastics into small hydrocarbons.28,29 We put forward “combined catalysts” (i.e., degradation catalyst and carbonization catalyst) to convert plastics into CNTs,30–34 in which the degradation catalyst promoted the decomposition of plastics into hydrocarbons and aromatics, and then the carbonization catalyst catalyzed the carbonization of the degradation products.

In contrast, the backbone chains of charring polymers generally undergo a series of chemical reactions such as decomposition, cyclization, aromatization and crosslinking, and gradually construct a carbon material frame.35–37 Evidently, the decomposition and crosslinking of charring polymers are critical for the formation of carbon materials. Compared to noncharring polymers, the “controlled carbonization” of charring polymers is more challenging since the decomposition and crosslinking occur simultaneously. The pre-crosslinking strategy is a typical method, in which charring polymers are crosslinked and the decomposition reaction is greatly inhibited when heating.38 Controlling the heating rate or temperature is also applied for the carbonization of charring polymers such as polyacrylonitrile.39 Nevertheless, these methods could not accurately adjust the decomposition and crosslinking, nor precisely modulate the porosity of carbon products, which is critical for their performance.

In this contribution, inspired by “combined catalysts” for noncharring polymers, we propose a novel strategy of molten salts (ZnCl2/NaCl) to achieve “controlled carbonization” of PET. PET is widely used as a packaging material for food and consumer products; for example, the number of annual worldwide PET beverage waste bottles is up to 62.8 billion. PET is a good carbon source to prepare carbon materials,40,41 but the “controlled carbonization” of PET has not been realized yet. During pyrolysis of PET, vinyl-terminated chain fragments are produced from ester bond scission, and aromatic rings are yielded from the decomposition of chain end groups or by ester–phthalic ring bond scission.42,43 The intermolecular condensation of vinyl-terminated chain fragments or interconnection of aromatic rings leads to the formation of a carbon material frame.44,45 Thereby, the key to the “controlled carbonization” of PET is to adjust the decomposition of PET into vinyl-terminated chain fragments and aromatic rings and simultaneously promote their crosslinking. Here, we prove that ZnCl2 catalyzes the dehydration and decarboxylation of PET to form vinyl-terminated chain fragments and aromatic rings, which facilitates the formation of a carbon material frame. Meanwhile, ZnCl2/NaCl eutectic salts act as porogens to produce mesopores and macropores. Thanks to the high specific surface area, fast water transportation channels, high solar absorption efficiency, and low thermal conductivity, the resultant hierarchically porous carbon displays excellent performance in the solar steam evaporation from seawater.

2. Experimental section

2.1 Materials

NaCl, ZnCl2, ethanol, hydrochloric acid, melamine, rhodamine B (RB), methyl blue (MB), and dimethylpolysiloxane were purchased from Sinopharm Chemical Reagent Co., Ltd. PET and polycarbonate were purchased from Shandong Usolf Co., Ltd. Waste PET beverage bottles were collected from a local supermarket (Wuhan), washed and cut into small pieces (ca. 2 mm). Carbon black, activated carbon, CNTs and carbon fiber were kindly provided by Nanjing XFNANO Materials Tech Co., Ltd. River water, wastewater and seawater were collected from Donghu Lake (Wuhan, Fig. S1 in the ESI), the Huxi river (Wuhan, Fig. S2), and the South Sea (near Hainan, Fig. S3), respectively.

2.2 Synthesis of HPC-x from PET

As shown in Fig. 1, a PET–ZnCl2/NaCl-x mixture was first prepared by mixing PET pellets (5.0 g) with a designated amount of ZnCl2/NaCl (=58/42, according to Fechler's work46), where x represents the mass ratio of ZnCl2/NaCl eutectic salts to PET. The addition of NaCl to ZnCl2 decreases the melting temperature of ZnCl2 from ca. 290 to ca. 250 °C, close to the melting temperature of PET (230–250 °C). This is the main reason for the choice of the ZnCl2/NaCl mixture here, instead of ZnCl2 or NaCl alone. Subsequently, PET–ZnCl2/NaCl-x was placed in a crucible, heated at 280 °C for 10 min, and calcined at 550 °C for 8 min. The heating rate was kept as 10 °C min−1. Our previous results displayed that compared to the cases of 450 and 650 °C, the carbon product prepared at 550 °C showed the highest yield and rich nanopores (Table S1), and thus the carbonization temperature was selected as 550 °C. After slowly cooling down to ambient temperature, the as-obtained carbon was washed with hydrochloric acid (0.5 mol L−1) and water until pH = 7, and then dried, and it was denoted as HPC-x (here HPC is the abbreviation of hierarchically porous carbon). The yield of HPC-x was calculated by dividing the amount of obtained carbon by that of PET in the mixture. For comparison, neat PET was carbonized in the absence of ZnCl2/NaCl to prepare HPC-0. Likewise, W-HPC-2 was obtained using waste PET as the carbon source at the mass ratio of ZnCl2/NaCl eutectic salts to PET of 2 (Fig. S4). Similarly, the carbonization of polycarbonate to prepare porous carbon was conducted in the presence of ZnCl2/NaCl, and N-doped porous carbon was obtained by the carbonization of PET in the presence of both melamine (10 wt%) and ZnCl2/NaCl.
image file: c9ta07663h-f1.tif
Fig. 1 Schematic illustration of the “controlled carbonization” of PET using ZnCl2/NaCl molten salts as catalysts and porogens to fabricate HPC for solar steam evaporation from seawater.

2.3 Characterization

The morphology of HPC-x was observed by means of a field-emission scanning electron microscope (FESEM, SU8010) with an accelerating voltage of 10 kV. The surface elemental distribution of HPC-x was analyzed using an energy dispersive X-ray spectrometer (EDX, Genesis 2000). The microstructure of HPC-x was investigated using a transmission electron microscope (TEM, Tecnai G2 F30) at an accelerating voltage of 100 kV and high-resolution TEM (HR-TEM) on a Tecnai G2 F30 field-transmission electron microscope operating at 200 kV. The porosity of HPC-x was measured based on N2 adsorption/desorption isotherms at 77 K using a surface area analyzer (Micromeritics ASAP 2460). The pore size distribution was obtained by the quenched solid density functional theory (DFT) or Barrett–Joyner–Halenda (BJH) method. The specific surface area (SBET) was calculated using the Brunauer–Emmett–Teller (BET) equation. The pore volume was calculated using the t-method. The Raman spectrum was collected using a confocal Raman microscope (inVia Reflex, excitation beam wavelength = 532 nm). The phase structure of HPC-x was analyzed by X-ray diffraction (XRD, SmartLab-SE) with Cu Kα radiation operating at 40 kV and 200 mA. The elemental content of HPC-x was determined using an elemental analyzer (Vario EL III). The surface elemental composition of HPC-x was characterized by means of X-ray photoelectron spectroscopy (XPS) carried out on a VG ESCALAB MK II spectrometer using Al Kα excitation radiation from an X-ray source operated at 10 kV and 10 mA. The thermal stability of HPC-x was measured by thermogravimetric analysis (TGA) using a TA Instruments SDT Q600 in air atmosphere with a heating rate of 10 °C min−1. The functional groups of HPC-x and PET samples (i.e., PET, PET–NaCl, and PET–ZnCl2/NaCl-2) pretreated at 280 °C for 10 min were characterized by Fourier transform infrared spectroscopy (BRUKER Vertex 80 FT-IR, Germany). The absorption spectra of HPC-0 and HPC-2 were recorded using a UV-vis-NIR spectrophotometer (Lambda 750 S) with an integrating sphere. The thermal conductivity of HPC-2 was measured using a thermal conductivity measuring instrument (Hot Disk, TPS 2500, Sweden; see Fig. S5 for detailed information).

2.4 Solar steam generation by HPC-x

A solar light simulator (CEL-S500L) was used for the steam generation experiment (Fig. S6). To prepare the HPC-x membrane, an optimized amount of HPC-x (2.8 mg) was added to 14 mL of ethanol and sonicated for 3 min. The dispersion was filtered onto a hydrophilic polyvinylidene fluoride (PVDF) membrane (diameter = 50 mm, pore size = 0.22 μm) using a vacuum pump. Afterwards, the as-synthesized HPC-x membrane was put on the surface of a 2 cm-thick polystyrene foam, which was used as the thermal insulating layer, and wrapped by hydrophilic silk to make sure that water could reach the membrane. The surface temperature of the membrane was monitored using an infrared thermal imaging camera (DM-I220, Dongmei). Infrared images were captured with this camera. The mass change of pure water was measured using an electronic balance (JA2003, Soptop). Seawater, river water, wastewater, oil-contaminated water (i.e., dimethylpolysiloxane with a concentration of 10[thin space (1/6-em)]000 ppm), and dye-containing water (i.e., RB with a concentration of 20 ppm and MB with a concentration of 20 ppm, respectively) were also used. The evaporation rate (m, kg m−2 h−1), solar-to-vapor conversion efficiency (η, %), and enhancement factor (E.F.) were calculated using eqn (1)–(3), respectively:
 
m = Δm/(S × t)(1)
 
η = m′ × hlv/Pin(2)
 
E.F. = m(absorber)/m(blank)(3)
where Δm is the mass change in 1 h (kg), S is the area of the HPC-x membrane (m2), m′ is the evaporation rate after subtracting the evaporation rate in the dark (kg m−2 h−1), t is the time of solar irradiation (1 h), hlv is the total enthalpy of the liquid–vapor phase transition of water including the sensible heat (kJ kg−1), and Pin is the incident light power on the solar absorber (kW m−2). Solar steam generation without absorbers was studied as a reference experiment, denoted as blank. UV-vis absorption spectra of the dye-containing water and the condensed water were recorded on a UV-6100 spectrometer (METASH). The metallic ion concentrations of the seawater and the condensed water were determined using an inductively coupled plasma-optical emission spectrometer (ICP-OES, JY 200-2, HORIBA Scientific). The optical images of the oil/water emulsion and the condensed water were observed by using an optical microscope (Mshot, MS60).

3. Results and discussion

3.1 Morphology and textural structure of HPC-x

First, ZnCl2/NaCl eutectic salts are found to drastically reduce the packing density of HPC-2 (Fig. 2a), reflecting that the eutectic salts influence the morphology of HPC-2. The morphologies of HPC-0 and HPC-2 were observed by SEM. HPC-0 is composed of many smooth-surfaced macroparticles with the size range of 50–200 μm (Fig. 2b and S7). In contrast, HPC-2 displays an irregular, inter-connected nanoparticle morphology, in which a number of nanopores in the size range of 20–90 nm are clearly observed (Fig. 2c and S7). The elemental mapping based on EDX spectroscopy portrays the homogeneous distribution of carbon and oxygen elements in HPC-2 (Fig. 2d–f).
image file: c9ta07663h-f2.tif
Fig. 2 (a) Photographs of HPC-0 and HPC-2. SEM images of (b) HPC-0 and (c and d) HPC-2. EDX maps of HPC-2 (e) for the C element and (f) for the O element according to (d) the SEM image. (g) N2 adsorption–desorption isotherms of HPC-0 and HPC-2 at 77 K. Pore size distribution plots of HPC-0 and HPC-2 obtained using (h) DFT and (i) BJH models.

The textural properties of HPC-0 and HPC-2 were investigated by N2 physisorption at 77 K (Fig. 2g). HPC-0 shows a type-I curve with a significant N2 uptake in the low-pressure range (P/P0 < 0.1), testifying the presence of many micropores.47 Interestingly, HPC-2 exhibits combined type I/IV physisorption isotherms with a type-H4 hysteresis loop at a relative pressure between 0.4 and 0.9, which corresponds to the filling and emptying of mesopores by capillary condensation. At high relative pressures (P/P0 > 0.9), the curve shows a sharp nitrogen uptake due to the existence of large mesopores and macropores.48 The pore size distribution is calculated using DFT (Fig. 2h) and BJH (Fig. 2i) models, respectively. The micropore size of HPC-2 is centered at 0.50 nm, which is smaller than that of HPC-0 (0.64 nm). Micropores are mainly derived from the intermolecular condensation of vinyl-terminated chain fragments and the interconnection of aromatic rings. The decreasing micropore size implies that ZnCl2/NaCl eutectic salts influence the decomposition and crosslinking reactions of PET. Furthermore, HPC-2 has a lot of mesopores and macropores in the size range of 2–100 nm. These nanopores are absent in HPC-0, which is in agreement with the SEM result (Fig. 2b). The BET surface area (SBET), specific surface area of micropores (Smicro), specific surface area of mesopores and macropores (Smeso+macro), and pore volume (V) are summarized in Table 1. HPC-2 shows an SBET of 776.2 m2 g−1, Smeso+macro of 244.7 m2 g−1, and V of 0.663 cm3 g−1, in comparison to HPC-0 (650.3 m2 g−1, 82.2 m2 g−1, and 0.443 cm3 g−1, respectively). Apparently, HPC-2 possesses 50% more pore volume than HPC-0 and significantly 197% more Smeso+macro.

Table 1 Yield, textural properties, elemental composition, evaporation rate, efficiency, and enhancement factor of HPC-x
Sample Salts/PET (mass ratio) Yield (wt%) S BET (m2 g−1) S micro (m2 g−1) S meso+macro (m2 g−1) V (cm3 g−1) C (%) H (%) O (%) Evaporation rate (kg m−2 h−1) Efficiency (%) Enhancement factor
a Waste PET beverage bottles were used as carbon sources.
HPC-0 0 16.2 650.3 568.1 82.2 0.443 94.4 3.1 2.5 0.79 38 1.27
HPC-0.5 0.5 15.7 812.7 580.2 232.5 0.530 88.8 1.6 9.6 1.28 71 2.06
HPC-1 1 21.4 827.6 552.9 274.7 0.653 89.8 1.9 8.3 1.59 92 2.56
HPC-2 2 24.9 776.2 531.5 244.7 0.663 89.1 1.8 9.1 1.68 97 2.71
HPC-5 5 24.9 495.1 395.4 99.7 0.547 84.2 1.9 13.9 1.46 83 2.35
HPC-10 10 17.1 476.5 328.4 148.2 0.525 78.6 1.9 19.5 1.45 82 2.34
W-HPC-2a 2 22.3 645.6 495.9 149.7 0.606 88.4 1.8 9.8 1.67 97 2.69


To gain more detailed information about the internal nanoporous structure of HPC-2, TEM and HRTEM observations were conducted. HPC-2 presents a rough surface and abundant large mesopores and macropores in the size range of ca. 10–80 nm (Fig. 3a and b). HRTEM images of HPC-2 indicate the presence of rich small mesopores of size around 2–5 nm and micropores around 0.4–2 nm (Fig. 3c and d), confirming the hierarchically porous architecture of HPC-2. Additionally, HPC-2 exhibits a slight degree of ordering with an interlayer spacing between graphitic layers of ca. 0.35 nm (Fig. 3d), which reveals a low graphitization degree.


image file: c9ta07663h-f3.tif
Fig. 3 (a and b) TEM and (c and d) HRTEM images of HPC-2. Nanopores are indicated with red circles.

3.2 Phase structure and elemental composition of HPC-x

XRD and Raman measurements were employed to examine the phase structures of HPC-0 and HPC-2. The two weak and broad diffraction peaks at 2θ = 17.1° and 43.9° for HPC-0 (or 2θ = 24.0° and 43.9° for HPC-2) are assigned to the graphitic (002) and (101) planes, respectively (Fig. 4a), according to the previous work.49 HPC-0 and HPC-2 thus display a low degree of graphitization. The (002) plane of HPC-2 shifts to a lower 2θ value compared to HPC-0, indicating that the interlayer spacing between graphitic layers of HPC-2 is smaller than that of HPC-0. This result suggests that ZnCl2/NaCl eutectic salts facilitate the relatively more compact arrangement of graphitic layers.
image file: c9ta07663h-f4.tif
Fig. 4 (a) XRD patterns, (b) Raman spectra, (c) TGA curves, and (d) derivative TGA curves of HPC-0 and HPC-2.

Fig. 4b displays the Raman spectra of HPC-0 and HPC-2. The peak at ca. 1588 cm−1 (G band) is due to an E2g mode of hexagonal graphite and related to the vibration of sp2-bonded carbon atoms in a graphite layer.50 The D band at ca. 1341 cm−1 is associated with the vibration of carbon atoms with dangling bonds in the plane termination of disordered graphite. The intensity ratio of the D band and G band provides information about the defects or disordered carbon.51 Raman curve fittings of D and G bands are shown in Fig. S8. Compared to HPC-0, the relatively larger ID/IG ratio of HPC-2 reflects more defects or disordered carbon, possibly due to the enhanced nanoporous structure of HPC-2 caused by ZnCl2/NaCl.

To study the purity and thermal stability of HPC-0 and HPC-2, TGA measurement in air atmosphere was conducted (Fig. 4c and d). The weight loss stage from 400 to 580 °C is due to the oxidation of the carbon framework. The residues of HPC-0 and HPC-2 at 600 °C are less than 0.8 wt%, indicating their high purity. HPC-2 shows the maximum oxidation temperature at 518 °C, which is lower than that of HPC-0 (547 °C). HPC-2 thus displays lower thermal stability owing to the presence of more nanopores, which is consistent with the results of SEM, TEM, XRD and Raman measurements.

To determine the functional groups of HPC-0 and HPC-2, FT-IR measurement was carried out (Fig. 5a). For HPC-0, the bands at 3646 and 3423 cm−1 are attributed to the –OH stretching vibration of water and hydroxyl (or carboxyl), respectively. The band at 3051 cm−1 is due to the [double bond, length as m-dash]C–H stretching vibration. The bands at 2917 and 2850 cm−1 are due to the –CH asymmetric and symmetric stretching vibrations, respectively. The band at 1700 cm−1 is ascribed to the –C[double bond, length as m-dash]O stretching vibration. The band at 1600 cm−1 is attributed to the stretching vibration of C[double bond, length as m-dash]C. The bands at 1471 and 1407 cm−1 are due to aromatic ring vibration. Furthermore, the bands at 1249 and 1094 cm−1 are ascribed to the (C[double bond, length as m-dash]O)–C stretching of ester and C–O vibration, respectively. Compared to HPC-0, the absorption bands of HPC-2 due to [double bond, length as m-dash]C–H, C[double bond, length as m-dash]O, –O–C[double bond, length as m-dash]O and –C–O decrease, while other bands show no obvious changes.


image file: c9ta07663h-f5.tif
Fig. 5 (a) FT-IR curves of HPC-0 and HPC-2. (b) XPS spectra of HPC-0 and HPC-2. High-resolution C 1s XPS spectra of (c) HPC-0 and (d) HPC-2.

Elemental analysis results (Table 1) show that the contents of C, H and O elements are 94.4%, 3.1% and 2.5% for HPC-0, respectively, and 89.1%, 1.8% and 9.1% for HPC-2, respectively. XPS was further used to investigate the surface elemental compositions of HPC-0 and HPC-2. The surfaces of HPC-0 and HPC-2 contain oxygen and carbon elements (Fig. 5b). To figure out the chemical component and oxidation state of the carbon element, high-resolution C 1s XPS spectra of HPC-0 and HPC-2 are curve-fitted into four individual peaks (Fig. 5c and d): graphitic carbon (284.6 eV), –C–OH (285.6–285.8 eV), –C[double bond, length as m-dash]O (286.9–287.2 eV), and –COOH (288.8–288.9 eV).52 Both HPC-0 and HPC-2 possess rich functional groups such as –C–OH, –C[double bond, length as m-dash]O and –COOH. Compared to HPC-2, HPC-0 has relatively more –C–OH (37.0%), –C[double bond, length as m-dash]O (16.5%) and –COOH (12.0%), possibly due to the partial oxidation of surface carbon with a low degree of graphitization during carbonization.

3.3 Mechanism of the “controlled carbonization” of PET using ZnCl2/NaCl

To elaborate on the mechanism of “controlled carbonization” of PET using ZnCl2/NaCl eutectic salts, PET, a PET–NaCl mixture and a PET–ZnCl2/NaCl-2 mixture were pretreated at 280 °C for 10 min. Fig. 6a presents the photographs of these samples before and after heating. The color of PET changes from white to light brown after heating, and the PET–NaCl mixture shows a similar change to PET, except that the color seems a little browner. Surprisingly, the color of the PET–ZnCl2/NaCl-2 mixture turns black, proving that ZnCl2/NaCl eutectic salts enhance the carbonization of PET, even at as low as 280 °C.
image file: c9ta07663h-f6.tif
Fig. 6 (a) Photographs and (b) FT-IR curves of PET, PET–NaCl mixture, and PET–ZnCl2/NaCl-2 mixture pretreated at 280 °C for 10 min. (c) Mechanism of the “controlled carbonization” of PET using ZnCl2/NaCl eutectic salts.

Subsequently, FT-IR measurement was conducted to analyze the variations of functional groups in these samples (Fig. 6b). For the neat PET after pretreating at 280 °C, the band at ca. 3428 cm−1 is attributed to O–H stretching vibration. The bands at 2965 and 2905 cm−1 are ascribed to the C–H asymmetric and symmetric stretching vibrations, respectively. The band at 1721 cm−1 is due to the stretching vibration of C[double bond, length as m-dash]O. The band around 1641 cm−1 suggests the existence of conjugated double bonds. The bands at 1576, 1506, 1474 and 1414 cm−1 are due to aromatic ring vibration. Furthermore, the bands at 1249 and 1094 cm−1 are ascribed to the (C[double bond, length as m-dash]O)–C stretching vibration of ester and C–O vibration, respectively. The band at 1013 cm−1 is attributed to the C–C vibration. The band at around 971 cm−1 is due to the CH2 rocking. The bands at 828 and 721 cm−1 are due to the [double bond, length as m-dash]C–H deformation and C[double bond, length as m-dash]O deformation of the benzene ring, respectively. The PET–NaCl mixture after heating at 280 °C displays a similar FT-IR spectrum to PET, suggesting the negligible influence of NaCl on the decomposition or carbonization of PET.

However, the FT-IR spectrum of the PET–ZnCl2/NaCl-2 mixture after the same pretreating shows significant changes compared to that of PET or the PET/NaCl mixture. Absorption bands which are attributed to C[double bond, length as m-dash]O (1721 cm−1), O[double bond, length as m-dash]C–O (1249 cm−1) and C–O (1094 cm−1) clearly decrease, while the bands due to O–H of H2O (3568 cm−1), [double bond, length as m-dash]C–H (3163 cm−1) and C[double bond, length as m-dash]C (1641 cm−1) significantly increase. These above results demonstrate that ZnCl2 catalyzes the dehydration and decarboxylation of PET degradation products to form vinyl-terminated chain fragments and aromatic rings (Fig. 6c). These intermediate degradation products construct a carbon material frame after further crosslinking and cyclization. Recently, Yu et al. also found that ZnCl2 could promote the dehydration and carbonization of phenolic resin to prepare porous carbon.53

Based on the above results, the mechanism of the “controlled carbonization” of PET using ZnCl2/NaCl eutectic salts is illustrated in Fig. 7. First, ZnCl2/NaCl eutectic salts are melted with PET at ca. 250 °C, which is a prerequisite for precisely controlling the carbonization of PET. Subsequently, ZnCl2 promotes the dehydration and decarboxylation of PET into vinyl-terminated chain fragments and aromatic rings at 280 °C. After further crosslinking and cyclization, the carbon material frame is gradually formed with many micropores at 550 °C. Meanwhile, the remaining eutectic salts act as porogens to produce mesopores and macropores corresponding to the size of salt clusters and their percolation structures. Finally, HPC consisting of hierarchical micro-/meso-/micropores is produced after water washing. Moreover, waste PET can be converted into hierarchically porous carbon with a high SBET of 645.6 m2 g−1 using ZnCl2/NaCl eutectic salts (Fig. S9 and Table 1).


image file: c9ta07663h-f7.tif
Fig. 7 Schematic illustration of the mechanism of “controlled carbonization” of PET using ZnCl2/NaCl eutectic salts.

It is worth mentioning that although molten salts have been widely used for the fabrication of porous carbons from biomass,54,55 this work is the first report on the utilization of molten salts for the “controlled carbonization” of waste polymers. Compared to the previous polymer carbonization methods such as combined catalysts, active templates, and pyrolysis-gasification,28–34 molten salts are low cost and easily recycled for the next carbon production. Unlike the production of hierarchically porous carbon from a SiO2 template or using chemical activation,56 which involved the use of aggressive chemicals, molten salts are nontoxic and readily removed. Besides, ZnCl2/NaCl molten salts can be applied for the carbonization of other polyesters or the preparation of heteroatom-doped porous carbon. For example, when polycarbonate is used as a carbon source, the porous carbon product produced using molten salts shows an SBET of 632 m2 g−1, greatly higher than that of the carbon product obtained from the carbonization of polycarbonate without molten salts (110 m2 g−1). When melamine and molten salts are added in the carbonization of PET, the produced N-doped porous carbon product exhibits a high nitrogen content of 8.6 wt% while exhibiting a large SBET of 902 m2 g−1.

3.4 Effect of the mass ratio of ZnCl2/NaCl eutectic salts to PET on the carbonization of PET

The effect of the mass ratio of ZnCl2/NaCl eutectic salts to PET on the yield, morphology and porous structure of HPC-x produced from the carbonization of PET was investigated. When the eutectic salts/PET mass ratio increases from 0.5 to 10, the packing density of HPC-x goes down gradually (Fig. 8), and more nanoparticles are found in HPC-x (Fig. S10). The yield of HPC-x first rises, reaches a maximum value (24.9 wt% for HPC-2), and finally decreases (Table 1). During pyrolysis of PET, the molten state of ZnCl2/NaCl eutectic salts creates an ionic liquid-confined space to prevent the fast decomposition, cracking, shrinking and sintering of PET degradation products, which favors the crosslinking of the intermediate degradation products to construct a carbon framework. However, when more ZnCl2/NaCl eutectic salts are added, the intermediate degradation products will be excessively diluted by eutectic salts, which inhibits their condensation, interconnection, cyclization and crosslinking reactions, and certainly affects the formation of the carbon framework. The resultant HPC-x products show a low degree of graphitization (Fig. S11).
image file: c9ta07663h-f8.tif
Fig. 8 Photographs of HPC-0, 0.5, 1, 2, 5 and 10 produced from PET by adding different amounts of ZnCl2/NaCl eutectic salts.

Like HPC-2, other HPC-x (x = 0.5, 1, 5 and 10) products show combined type I/IV physisorption isotherms with a type-H4 hysteresis loop (Fig. 9a), suggesting the presence of both micropores and mesopores. From the pore size distribution plots calculated using DFT and BJH models (Fig. 9b and c), it can be observed that the micropores of HPC-x first decrease slightly from HPC-0.5 to HPC-2, and then drop obviously for HPC-5 and HPC-10. Small mesopores (2–10 nm) decrease from HPC-0.5 to HPC-2. Large mesopores and macropores increase for HPC-5 and HPC-10, and the size of macropores becomes larger. The comparison of SBET, Smicro, and Smeso+macro of HPC-x is summarized in Fig. 9d and Table 1. The Smicro of HPC-0.5, 1 and 2 is close to that of HPC-0, but the Smeso+macro of HPC-0.5, 1 and 2 is remarkably higher than that of HPC-0. The significant enhancement of Smeso+macro confirms the porogen role of ZnCl2/NaCl in producing nanopores. The Smicro of HPC-5 and 10 is much lower than that of HPC-0.5, 1 and 2. It is supposed that the high amount of eutectic salts affects the formation of micropores, probably because the excess amount of salts inhibits the crosslinking of PET degradation products as explained above. Nevertheless, the combination of the hierarchically porous structure and rich oxygen-containing functional groups such as –COOH and –OH (Fig. S12) undoubtedly favors the diffusion and transportation of water molecules in HPC-x.


image file: c9ta07663h-f9.tif
Fig. 9 (a) N2 adsorption–desorption isotherms of HPC-x at 77 K. Pore size distribution plots obtained using (b) DFT and (c) BJH models. (d) Comparison of SBET, Smicro and Smeso+macro of HPC-x.

Based on the above results, the mechanism of the effect of ZnCl2/NaCl eutectic salts/PET mass ratio on the carbonization of PET is proposed (Fig. 10). For neat PET, the shrinking and sintering of PET degradation products lead to the formation of bulk carbon possessing many micropores. In the presence of a low amount of eutectic salts (eutectic salts/PET = 0.5, 1 or 2), ZnCl2 enhances the crosslinking reaction of PET degradation products to improve the yield, meanwhile ZnCl2/NaCl2 eutectic salts act as a template to produce a lot of mesopores and macropores. When adding more eutectic salts (eutectic salts/PET = 5 or 10), the eutectic salts dilute PET degradation products and inhibit their crosslinking reactions, leading to the decreases of yield and micropores; meanwhile the aggregates of eutectic salts promote the formation of large-sized macropores.


image file: c9ta07663h-f10.tif
Fig. 10 Schematic representation of PET carbonization by adding different amounts of ZnCl2/NaCl eutectic salts.

3.5 Solar steam generation by HPC-x

Ideal photothermal materials should show broadband sunlight absorbability, open porosity for rapid water transportation, low thermal conductivity, and high energy conversion efficiency. First, the optical absorption of HPC-2 in the visible and infrared regions was measured using a UV-vis-NIR spectrometer (Fig. 11a). HPC-2 shows relatively broad-band absorption of solar radiation with the optical absorption around 98–99%, higher than that of HPC-0 (ca. 74–97%). The enhanced light absorption efficiency is due to the improved porous structure which facilitates the light absorption via multiple scattering of absorbed light. The high light absorption efficiency is favorable for achieving better photothermal performance.57–59 Moreover, the low thermal conductivity of HPC-2 (0.117 W m−1 K−1) can reduce the conductive heat loss. In this regard, the surface temperature of HPC-0 and HPC-2 under solar light irradiation was recorded. Under 1 kW m−2 solar light irradiation, the surface temperature of HPC-2 rapidly rises to ca. 70 °C and this temperature is maintained steadily upon extending the irradiation time (Fig. 11b and c). However, HPC-0 and PVDF show a relatively low surface temperature of ca. 50 and 45 °C, respectively (Fig. 11b and S13). The above experimental results demonstrate the good performance of HPC-2 in the conversion of solar energy to thermal energy.
image file: c9ta07663h-f11.tif
Fig. 11 (a) UV-vis-NIR absorption spectra of HPC-0 and HPC-2. (b) Surface temperature of PVDF, HPC-0 and HPC-2 under 1 kW m−2 solar light irradiation. (c) Infrared images of HPC-2.

Taking advantage of the highly porous structure, excellent optical absorption, and low thermal conductivity, HPC-2 provides a promising platform for solar steam generation. To confirm this, the water evaporation performance of HPC-2 under 1 kW m−2 solar light irradiation was systematically investigated by using a lab-made, real-time measurement system. The mass change of water due to the steam generation was measured using an electronic analytical balance (Fig. 12a and Video S1). The water mass decreases approximately linearly with solar light irradiation time. After 60 min of solar light irradiation, the mass decrease of water in the presence of HPC-2 is 1.68 kg m−2 h−1, higher than that of pure water in the dark (0.21 kg m−2 h−1) or pure water under solar light irradiation (0.62 kg m−2 h−1). In comparison, the mass decreases of water with other samples under solar light irradiation are 0.71 kg m−2 h−1 for the PVDF membrane and 0.79 kg m−2 h−1 for HPC-0 (Fig. 12b). The solar-to-vapor conversion efficiency and enhancement factor of HPC-2 are 97% and 2.71, respectively, which are remarkably higher than those of HPC-0 (38% and 1.27, respectively) or the PVDF membrane (35% and 1.15, respectively). Furthermore, under the same experimental conditions, HPC-2 is better than commercial carbon materials such as activated carbon (1.44 kg m−2 h−1), CNTs (1.55 kg m−2 h−1), and carbon fiber (1.04 kg m−2 h−1). More importantly, HPC-2 outperforms the state-of-the-art carbon-based photothermal materials,60–62 in terms of evaporation rate and solar-to-vapor conversion efficiency (Fig. S14 and Table S2). These experimental results reveal that HPC-2 has superior performance in solar energy-driven steam generation. It is noted that the evaporation rate of W-HPC-2 is 1.67 kg m−2 h−1 (Fig. S15 and Table 1), indicating that the hierarchically porous carbon from waste PET also functions well in solar steam generation.


image file: c9ta07663h-f12.tif
Fig. 12 (a) Cumulative mass changes of water over time under various conditions: water in the dark (dark); water under 1 kW m−2 solar light irradiation (blank); water with the PVDF membrane under 1 kW m−2 solar light irradiation (PVDF); water with HPC-0 under 1 kW m−2 solar light irradiation (HPC-0); water with HPC-2 under 1 kW m−2 solar light irradiation (HPC-2). The inset in (a) shows the photograph of solar steam generation by HPC-2. (b) Solar-to-vapor conversion efficiency and evaporation rate obtained using PVDF, HPC-0 and HPC-2 under 1 kW m−2 solar light irradiation.

To clarify the influence of the morphology and textural properties of HPC-x on the performance of solar steam generation, the cumulative mass changes of water over time under 1 kW m−2 solar light irradiation of HPC-0.5, 1, 5 and 10 were also measured (Fig. S16a), and the evaporation rates were compared along with their SBET (Fig. S16b). HPC-5 and HPC-10 show much lower SBET than HPC-0, but their evaporation rates (1.46 and 1.45 kg m−2 h−1, respectively) are greatly higher than that of HPC-0 (0.79 kg m−2 h−1). Evidently, compared to bulk carbon (HPC-0), carbon nanoparticles of HPC-5 and HPC-10 facilitate the formation of a uniform carbon membrane on the PVDF membrane through vacuum filtration (Fig. S16c), which enhances the light absorption and the surface temperature of the membrane (Fig. S17). Meanwhile, for HPC-5 and HPC-10, the small thickness of carbon nanoparticles combined with a hierarchically porous structure enables fast diffusion kinetics of water and a high utilization degree of the overall porosity and surface area. They are also the reasons for the better performance of HPC-2 (1.68 kg m−2 h−1) than HPC-0.5 (1.28 kg m−2 h−1) or HPC-1 (1.59 kg m−2 h−1). Besides, HPC-5 and HPC-10 membranes seem to be as uniform as HPC-2, but HPC-2 shows a greatly higher evaporation rate than HPC-5 and HPC-10, proving that SBET is important in solar steam generation. High SBET of porous carbon provides more channels for the transportation of water molecules. Based on the above results, it is concluded that the nanostructure and high SBET of carbon materials are vital to achieve high performance in solar steam generation.

The solar steam generation of HPC-2 using organic dye-containing water (MB and RhB), seawater, river water, wastewater, and oil water was also investigated. First, the condensed water from organic dye-containing water under solar light irradiation is colorless to the naked eye (Fig. 13a). According to the UV-vis spectrum of the condensed water, the dye removal efficiency by HPC-2 is over 99.9%. Second, the Na+, K+, Mg2+ and Ca2+ concentrations of the condensed water from seawater dramatically decrease from ca. 1000–20[thin space (1/6-em)]000 ppm to ca. 0.3–10 ppm (Fig. 13b). More importantly, the collected fresh water satisfies the standards of healthy drinkable water as defined by the World Health Organization (WHO) and the U.S. Environmental Protection Agency (EPA). Furthermore, no oil droplets are found in the condensed water sample from the oil/water emulsion (Fig. 13c and d). The absence of the Tyndall effect confirms the high purity of the condensed water, which is different from the oil/water emulsion (the insets in Fig. 13c and d). Overall, when using river water, wastewater, seawater, and oil/water emulsion, the evaporation rates of HPC-2 show no obvious changes (Fig. 13e), and the collected fresh water is safe and of high purity. Finally, to evaluate the stability of HPC-2, the solar steam generation using seawater is repeated for 10 cycles, and the evaporation rate remains ca. 1.7 kg m−2 h−1 (Fig. 13f), suggesting the good stability of HPC-2.


image file: c9ta07663h-f13.tif
Fig. 13 (a) UV-vis absorption spectra before and after evaporation under 1 kW m−2. (b) Comparison of the metallic ion concentration in seawater and the condensed water; the dashed lines indicate the WHO and EPA standards for healthy drinkable water. Optical microscopy images of oil/water emulsion (c) before and (d) after purification. (e) Evaporation rates of different water sources when using HPC-2. (f) Recyclability of HPC-2 in seawater evaporation for 10 days at 1 kW m−2 for 1 h every day.

4. Conclusions

In summary, we have reported the “controlled carbonization” of PET to prepare hierarchically porous carbon by using ZnCl2/NaCl eutectic salts as both catalysts and porogens at 550 °C. Hierarchically porous carbon displays an irregular, inter-connected nanoparticle morphology with a high specific surface area and abundant oxygen-containing groups. ZnCl2 catalyzes the dehydration and decarboxylation of PET to form vinyl-terminated chain fragments and aromatic rings. These intermediate degradation products construct a carbon material frame with rich micropores. Meanwhile, ZnCl2/NaCl eutectic salts act as porogens to produce mesopores and macropores. The hierarchically porous carbon shows a high evaporation rate of 1.68 kg m−2 h−1, a high solar-to-vapor conversion efficiency of 97%, and good recyclability. The nanostructure and high specific surface area of carbon materials are crucial for achieving high performance in solar steam generation. Hierarchically porous carbon also works well using dye-containing water, wastewater, river water, seawater, and oil/water emulsion. The metallic ion removal efficiency of seawater is ca. 99.9%, and the dye removal efficiency is >99.9%. More importantly, it outperforms the state-of-the-art carbon-based photothermal materials in solar energy-driven steam generation. This work not only contributes to the field of seawater desalination, but also proposes a novel sustainable approach to convert waste polyesters into high value-added carbon nanomaterials. The applications of the “molten salts” strategy in the carbonization of other polyesters to prepare porous carbon and in the synthesis of heteroatom-doped porous carbon are being conducted in our laboratory. We hope this work can inspire the research of “controlled carbonization” of low-cost waste polymers to produce sustainable carbon materials for a wide variety of applications such as environmental remediation, and energy storage and conversion.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

We would like to thank the reviewers for their kind and valuable comments. J. G. thanks the National Natural Science Foundation of China (No. 51903099), and is also thankful for the Initiatory Financial Support from the Huazhong University of Science and Technology (No. 3004013134), and the opening fund of Hubei Key Laboratory of Materials Chemistry and Service Failure (No. 2019MCF01). Q. Z. thanks the 1000 Young Talent program and the Huazhong University of Science and Technology (No. 3004013118).

Notes and references

  1. P. Tao, G. Ni, C. Song, W. Shang, J. Wu, J. Zhu, G. Chen and T. Deng, Nat. Energy, 2018, 3, 1031–1041 CrossRef.
  2. M. Gao, L. Zhu, C. K. Peh and G. W. Ho, Energy Environ. Sci., 2019, 12, 841–864 RSC.
  3. G. Ni, G. Li, S. V. Boriskina, H. Li, W. Yang, T. Zhang and G. Chen, Nat. Energy, 2016, 1, 16126 CrossRef CAS.
  4. Y. Wu, S. Han, Y. Huang, Y. Shi and B. Zhang, J. Mater. Chem. A, 2018, 6, 18426–18429 RSC.
  5. D. Liu, C. Wang, Y. Yu, B.-H. Zhao, W. Wang, Y. Du and B. Zhang, Chem, 2019, 5, 376–389 CAS.
  6. L. Zhou, Y. Tan, J. Wang, W. Xu, Y. Yuan, W. Cai, S. Zhu and J. Zhu, Nat. Photonics, 2016, 10, 393–399 CrossRef CAS.
  7. D. Wu, D. Qu, W. Jiang, G. Chen, L. An, C. Zhuang and Z. Sun, J. Mater. Chem. A, 2019, 7, 8485–8490 RSC.
  8. X. Zhang, X. Wang, W. D. Wu, X. D. Chen and Z. Wu, J. Mater. Chem. A, 2019, 7, 6963–6971 RSC.
  9. J. Wang, Y. Li, L. Deng, N. Wei, Y. Weng, S. Dong, D. Qi, J. Qiu, X. Chen and T. Wu, Adv. Mater., 2017, 29, 1603730 CrossRef PubMed.
  10. N. Xu, X. Hu, W. Xu, X. Li, L. Zhou, S. Zhu and J. Zhu, Adv. Mater., 2017, 29, 1606762 CrossRef PubMed.
  11. J. Liu, Q. Liu, D. Ma, Y. Yuan, J. Yao, W. Zhang, H. Su, Y. Su, J. Gu and D. Zhang, J. Mater. Chem. A, 2019, 7, 9034–9039 RSC.
  12. F. Zhao, X. Zhou, Y. Shi, X. Qian, M. Alexander, X. Zhao, S. Mendez, R. Yang, L. Qu and G. Yu, Nat. Nanotechnol., 2018, 13, 489–495 CrossRef CAS PubMed.
  13. P. Mu, W. Bai, Y. Fan, Z. Zhang, H. Sun, Z. Zhu, W. Liang and A. Li, J. Mater. Chem. A, 2019, 7, 9673–9679 RSC.
  14. T. Li, Q. Fang, X. Xi, Y. Chen and F. Liu, J. Mater. Chem. A, 2019, 7, 586–593 RSC.
  15. Y. Li, X. Cui, M. Zhao, Y. Xu, L. Chen, Z. Cao, S. Yang and Y. Wang, J. Mater. Chem. A, 2019, 7, 704–710 RSC.
  16. Y. Li, T. Gao, Z. Yang, C. Chen, W. Luo, J. Song, E. Hitz, C. Jia, Y. Zhou, B. Liu, B. Yang and L. Hu, Adv. Mater., 2017, 29, 1700981 CrossRef PubMed.
  17. Y. Chen, Y. Shi, H. Kou, D. Liu, Y. Huang, Z. Chen and B. Zhang, ACS Sustainable Chem. Eng., 2019, 7, 2911–2915 CrossRef CAS.
  18. P. J. Kim, H. D. Fontecha, K. Kim and V. G. Pol, ACS Appl. Mater. Interfaces, 2018, 10, 14827–14834 CrossRef CAS PubMed.
  19. J. R. Jambeck, R. Geyer, C. Wilcox, T. R. Siegler, M. Perryman, A. Andrady, R. Narayan and K. L. Law, Science, 2015, 341, 768–771 CrossRef PubMed.
  20. R. Geyer, J. R. Jambeck and K. L. Law, Sci. Adv., 2017, 3, e1700782 CrossRef PubMed.
  21. B. Alireza and M. Gordon, Chem. Eng. J., 2012, 195–196, 377–391 Search PubMed.
  22. C. Zhuo, B. Hall, H. Richter and Y. Levendis, Carbon, 2010, 48, 4024–4034 CrossRef CAS.
  23. C. Zhuo and Y. A. Levendis, J. Appl. Polym. Sci., 2014, 131, 39931–39944 CrossRef.
  24. H. Zhang, X.-L. Zhou, L.-M. Shao, F. Lü and P.-J. He, ACS Sustainable Chem. Eng., 2019, 7, 3801–3810 CrossRef CAS.
  25. S. Natarajan, H. C. Bajaj and V. Aravindan, J. Mater. Chem. A, 2019, 7, 3244–3252 RSC.
  26. Z. Yang, Q. Zhang, G. Luo, J.-Q. Huang, M.-Q. Zhao and F. Wei, Appl. Phys. A, 2010, 100, 533–540 CrossRef CAS.
  27. J. Gong, X. Chen and T. Tang, Prog. Polym. Sci., 2019, 94, 1–32 CrossRef CAS.
  28. C. Wu, M. A. Nahil, N. Miskolczi, J. Huang and P. T. Williams, Environ. Sci. Technol., 2014, 48, 819–826 CrossRef CAS PubMed.
  29. D. Yao, Y. Zhang, P. T. Williams, H. Yang and H. Chen, Appl. Catal., B, 2018, 221, 584–597 CrossRef CAS.
  30. T. Tang, X. Chen, X. Meng, H. Chen and Y. Ding, Angew. Chem., Int. Ed., 2005, 44, 1517–1520 CrossRef CAS PubMed.
  31. J. Gong, J. Liu, L. Ma, X. Wen, X. Chen, D. Wan, H. Yu, Z. Jiang, E. Borowiak-Palen and T. Tang, Appl. Catal., B, 2012, 117–118, 185–193 CrossRef CAS.
  32. J. Gong, J. Liu, Z. Jiang, J. Feng, X. Chen, L. Wang, E. Mijowska, X. Wen and T. Tang, Appl. Catal., B, 2014, 147, 592–601 CrossRef CAS.
  33. J. Gong, J. Liu, Z. Jiang, X. Chen, X. Wen, E. Mijowska and T. Tang, Appl. Catal., B, 2014, 152–153, 289–299 CrossRef CAS.
  34. J. Gong, J. Feng, J. Liu, Z. Jiang, X. Chen, E. Mijowska, X. Wen and T. Tang, Chem. Eng. J., 2014, 248, 27–40 CrossRef CAS.
  35. Z. Li, D. Wu, X. Huang, J. Ma, H. Liu, Y. Liang, R. Fu and K. Matyjaszewski, Energy Environ. Sci., 2014, 7, 3006–3012 RSC.
  36. J.-S. M. Lee, M. E. Briggs, T. Hasell and A. I. Cooper, Adv. Mater., 2016, 28, 9804–9810 CrossRef CAS PubMed.
  37. H.-B. Zhao, B.-W. Liu, X.-L. Wang, L. Chen, X.-L. Wang and Y.-Z. Wang, Polymer, 2014, 55, 2394–2403 CrossRef CAS.
  38. H. Liu, S. Li, H. Yang, S. Liu, L. Chen, Z. Tang, R. Fu and D. Wu, Adv. Mater., 2017, 29, 1700723 CrossRef PubMed.
  39. M. Kopeć, M. Lamson, R. Yuan, C. Tang, M. Kruk, M. Zhong, K. Matyjaszewski and T. Kowalewski, Prog. Polym. Sci., 2019, 92, 89–134 CrossRef.
  40. L. Cui, X. Wang, N. Chen, B. Ji and L. Qu, Nanoscale, 2017, 9, 9089–9094 RSC.
  41. L. Z. Wei, N. Yan and Q. W. Chen, Environ. Sci. Technol., 2011, 45, 534–539 CrossRef CAS PubMed.
  42. B. J. Holland and J. N. Hay, Polymer, 2002, 43, 1835–1847 CrossRef CAS.
  43. S. V. Levchik and E. D. Weil, Polym. Adv. Technol., 2004, 15, 691–700 CrossRef CAS.
  44. J. Zheng, P. Cui, X. Tian and K. Zheng, J. Appl. Polym. Sci., 2007, 104, 9–14 CrossRef CAS.
  45. K. Kawai, H. Kondo and H. Ohtani, Polym. Degrad. Stab., 2008, 93, 1781–1785 CrossRef CAS.
  46. N. Fechler, T.-P. Fellinger and M. Antonietti, Adv. Mater., 2013, 25, 75–79 CrossRef CAS PubMed.
  47. J. Gong, M. Antonietti and J. Yuan, Angew. Chem., Int. Ed., 2017, 56, 7557–7563 CrossRef CAS PubMed.
  48. N. Díez, M. Qiao, J. L. Gómez-Urbano, C. Botas, D. Carriazo and M. M. Titirici, J. Mater. Chem. A, 2019, 7, 6126–6133 RSC.
  49. Z. Q. Li, C. J. Lu, Z. P. Xia, Y. Zhou and Z. Luo, Carbon, 2007, 45, 1686–1695 CrossRef CAS.
  50. J. Gong, H. Lin, M. Antonietti and J. Yuan, J. Mater. Chem. A, 2016, 4, 7313–7321 RSC.
  51. A. Panahi, Z. Wei, G. Song and Y. A. Levendis, Ind. Eng. Chem. Res., 2019, 58, 3009–3023 CrossRef CAS.
  52. L. Su, Z. Zhou, X. Qin, Q. Tang, D. Wu and P. Shen, Nano Energy, 2013, 2, 276–282 CrossRef CAS.
  53. Z.-L. Yu, G.-C. Li, N. Fechler, N. Yang, Z.-Y. Ma, X. Wang, M. Antonietti and S.-H. Yu, Angew. Chem., Int. Ed., 2016, 55, 14623–14627 CrossRef CAS PubMed.
  54. J. Pampel, A. Mehmood, M. Antonietti and T. P. Fellinger, Mater. Horiz., 2017, 4, 493–501 RSC.
  55. J. Pampel and T.-P. Fellinger, Adv. Energy Mater., 2016, 6, 1502389 CrossRef.
  56. J. Gong, J. Liu, X. Chen, Z. Jiang, X. Wen, E. Mijowska and T. Tang, J. Mater. Chem. A, 2015, 3, 341–351 RSC.
  57. R. Hu, J. Zhang, Y. Kuang, K. Wang, X. Cai, Z. Fang, W. Huang, G. Chen and Z. Wang, J. Mater. Chem. A, 2019, 7, 15333–15340 RSC.
  58. Q. Zhang, H. Yang, X. Xiao, H. Wang, L. Yan, Z. Shi, Y. Chen, W. Xu and X. Wang, J. Mater. Chem. A, 2019, 7, 14620–14628 RSC.
  59. R. Chen, X. Wang, Q. Gan, T. Zhang, K. Zhu and M. Ye, J. Mater. Chem. A, 2019, 7, 11177–11185 RSC.
  60. Y. Ito, Y. Tanabe, J. Han, T. Fujita, K. Tanigaki and M. Chen, Adv. Mater., 2015, 27, 4302–4307 CrossRef CAS PubMed.
  61. L. Zhou, Y. Tan, D. Ji, B. Zhu, P. Zhang, J. Xu, Q. Gan, Z. Yu and J. Zhu, Sci. Adv., 2016, 2, e1501227 CrossRef PubMed.
  62. P. Mu, Z. Zhang, W. Bai, J. He, H. Sun, Z. Zhu, W. Liang and A. Li, Adv. Energy Mater., 2019, 9, 1802158 CrossRef.

Footnote

Electronic supplementary information (ESI) available: Yield and SBET of carbon, comparison of HPC-2 with other materials in solar steam generation, photographs of wastewater and the condensed water, the photograph of W-HPC-2, SEM images of HPC-x, fitting results of Raman spectra of HPC-x, N2 adsorption–desorption isotherms of W-HPC-2, XRD patterns, FT-IR curves and XPS spectra of HPC-x, infrared images of HPC-0 and PVDF, applications of HPC-x and W-HPC-2 in solar steam generation, and surface temperature of HPC-x. See DOI: 10.1039/c9ta07663h

This journal is © The Royal Society of Chemistry 2019